INDEX

Explanations

Roman numerals IX (9) and X (10)

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

--------------------------------------------------------

-0.71

 rule

-0.65

 clutch

-0.63

TPS

-0.62

 mount

-0.58

fixes

-0.58

thritis

-0.58

xon

-0.58

 Serious

-0.57

forth

-0.56

POSITIVE LOGITS

iew

1.10

irus

1.09

isions

1.09

olution

1.06

isible

1.03

entric

1.00

ision

1.00

itamin

0.97

iral

0.97

orst

0.97

Activations Density 0.030%

Roman numerals IX (9) and X (10)

No Comments

No Known Activations